Retrospective data suggest NK cells play a role in protecting recipients from graft versus host disease (GVHD) in the setting of killer IgG-like receptor (KIR) ligand incompatibility. In humans, this protective effect is most evident with MHC mismatched transplantation, usually following in vivo or in vitro T-cell depletion. In MHC mismatched murine transplant models, lethal GVHD is reduced following the adoptive infusion of KIR ligand mismatched NK cells; it is unknown whether NK cells can mediate similar protective effects following MHC matched transplantation. Therefore, we investigated the impact of adoptively infusing KIR ligand mismatched NK cells on GVHD in an MHC matched T-cell replete murine model of allogeneic transplantation. Balb/C recipient mice underwent allogeneic bone marrow (8 x 106 cells) and splenocyte (15 x 106 cells) transplantation from B10.d2 donors following 950cGy of irradiation. Allogeneic B10.d2 donor NK cells were first isolated by negative depletion using magnetic beads selecting for CD4, CD5, CD8a, CD19, Gr-1 and Ter-119, and then expanded over 4-6 days in vitro in DMEM media containing 10% FCS and 500U/ml of IL-2. NK cell subsets (KIR ligand matched vs. KIR ligand mismatched) were then isolated by flow cytometry into Ly49I/C+ NK cells (KIR ligand mismatched in the GVHD direction for Balb/C recipients) and Ly49A/G+ NK cells (KIR ligand matched for Balb/C recipients). On day +4, recipient mice received a single tail vein injection with either KIR ligand matched, KIR ligand mismatched or unsorted “bulk” NK cells (0.5–1.0 x 106 NK cells). All (9/9) control transplant recipients (no adoptive NK cell infusion) as well as recipients of Ly49A/G (KIR ligand matched) NK cells (13/13) developed skin GVHD, in contrast to 4/7 (57%, p=0.03) recipients of bulk NK cells and only a minority (13% [1/8], p < 0.01) of animals receiving KIR ligand mismatched NK cells. Using a cumulative clinical GVHD scoring system (total score = 9), overall GVHD was decreased in recipients of KIR ligand mismatched NK cells (median score = 0 at day +45) compared to mice that received KIR ligand matched NK cells (median score = 3; p = 0.15) or no NK cells (median score = 3; p= 0.12); no significant difference in survival was observed between cohorts. This murine model provides the first in vivo evidence that adoptively infused KIR ligand mismatched allogeneic NK cells reduce GVHD following T-cell replete MHC matched allogeneic transplantation. The impact of infusing multiple doses of KIR ligand mismatched NK cells on GVHD and their ability to induce a graft-vs-tumor effect in tumor bearing Balb/c mice is currently being evaluated.

Author notes

Corresponding author

Sign in via your Institution